Quartiles and Mel Frequency Cepstral Coefficients vectors in Hidden Markov-Gaussian Mixture Models classification of merged heart sounds and lung sounds signals.
Pedro MayorgaDaniela IbarraVesna ZeljkovicChristopher DruzgalskiPublished in: HPCS (2015)
Keyphrases
- mel frequency cepstral coefficients
- acoustic features
- spectral features
- gaussian mixture model
- feature vectors
- spectral analysis
- mixture model
- linear predictive
- speaker identification
- speaker recognition
- speech signal
- speaker verification
- hyperspectral images
- automatic speech recognition
- visual features
- remote sensing images
- eeg data
- structural features
- sound source
- audio signal
- music information retrieval
- noisy environments
- cepstral coefficients
- feature space
- feature set
- audio features
- em algorithm
- cross correlation
- speech recognition
- expectation maximization
- principal component analysis
- feature extraction
- machine learning
- eeg signals
- unsupervised learning
- text classification
- feature selection